Tracking on the Field

Racial Diversity and Disparity in Football Quarterbacking

Authors

Affiliations

Magdalena Bennett
Principal Investigator

The University of Texas at Austin

RESEARCH ASSISTANTS

Shishir Nambi

The University of Texas at Austin

Benny Wang

The University of Texas at Austin

Maryam Sadiq

The University of Texas at Austin

Abstract

This study explores the racial imbalances present at various tiers within American competitive football, including high school, college, and the National Football League (NFL), with a focus on the quarterback (QB) position, a critical role within the team. We examine the selection processes at each level, shedding light on the recruitment and drafting patterns. Our initial findings reveal significant disparities in the recruitment probabilities for QBs based on race: Highly skilled high school QBs of color have a lower chance of being recruited for college teams of 12 percentage points in comparison to their white peers, even when accounting for physical attributes and playing techniques. Furthermore, elite college QBs of color face a 5 percentage point lower probability of entering the NFL relative to white players. These discrepancies remain even after adjustments to equate the playing styles of black QBs to those of white QBs.

Introduction

The intersection of race and the quarterback position in American football has been a topic of considerable discussion and analysis over the years. Historically, the quarterback position in the National Football League (NFL) was predominantly occupied by white athletes, reflecting broader societal attitudes and stereotypes about leadership and intellectual roles in sports. However, the landscape has been changing gradually.

Recent years have witnessed a significant shift in this trend. As of 2023, the NFL has seen a notable increase in the number of quarterbacks of African American and other racial backgrounds leading teams, challenging long-standing biases. For instance, at the start of the 2023 season, approximately 44% of starting quarterbacks in the NFL were African American, a record for the league and a marked increase from just 25% two decades earlier. This change not only reflects a shift in the racial dynamics within the sport but also indicates a broader societal progression towards inclusivity and recognition of diverse talents.

This study aims to delve deeper into the statistical trends of potential racial discrimination in the quarterback position, analyzing selection at different stages of a player’s career: from high school to college, and, potentially, from college to pro-ball. In particular, we will analyze shifts and selections at these two stages, shown in the following plot:

Figure 1. Selection process for Quarterbacks at different stages by race (Source: Own elaboration based on ESPN Top High School QB data 2006-2017)

Data

Sources

To analyze the progression of racial disparity in the quarterback position, we merge several data sources:

1. Player Stats for College Football (1980 - 2019): This dataset contains data from sports-reference.com about receiving players that play college football, including some key statistics such as number of games, passing yards, touchdowns, etc. Statistics from 2017 to 2019 were downloaded directly from the website. We also include data for all other players between 2000 and 2019 from each player’s individual page.

2. College Football Total QBR (2004-2019): Data set from espn.com, including 135 QBs per year, their ranking, and other statistics.

3. ESPN Top Quarterbacks (2006 - 2017): Data set from espn.com, which includes the top 100 recruits by position (QBs) each year. This dataset includes information about their college selection (if any), weight, height, high school, grade, and ranking.

4. College Football Stats (2006-2019): Data set from espn.com, with information about passing, rushing, and receiving leader boards for each year. It also includes specific statistics for each board, such as passing/rushing yards, touchdowns, and interceptions, among others. These datasets are also used to identify the position of each player in college. As a reference, the passing leader board includes around 500 players per year (mostly QBs), while the rushing and receiving leader board include over 1200 players each.

5. College Football Team Ranking (2005-2019): Information at the college team level for each year, including the Football Power Index metric from ESPN, as well as the record of wins and losses for the season.

6. Player information for College Football (1980 - 2022): This is a dataset gathered by a Reddit user (see here), including information about their position, statistics, school and town of origin, and some sparse NFL statistics as well.

7. Draft picks (1980-2022): Dataset with all draft picks by year and position.

8. 247 Sports data for top high school quarterbacks (2000 - 2017): Information for top high school quarterbacks each year (according to 247 Sports ranking), for both pro-style and dual threat QBs.

9. Perceived race classification (2006 - 2019): This is a variable manually gathered for QBs listed in the first and third data source. This is a perceived race classification based on images of the players, with three distinct categories: White, Black, and Other. Even though this variable does not capture the actual race of the player, it will be a proxy for race perception and how this aspect affects their selection at different stages.

We are currently working on adding high school demographic information for each player, in addition to extending the perceived race classification for additional years.

Data Description

High School Level

At the high school level, we are working with the top Quarterback recruits according to ESPN (100 to 150 each year from 2006) in addition to the top QB recruits from 247Sports (from 2000 onward), to analyze their potential path into college football. In total, we have 2140 unique high school QBs between 2000 and 2017. The following table shows the characteristics that are available in the data.

Table 1: Variable description and distribution for Top HS QBs (Source: ESPN and 247Sports)

College Level

In total, we have 3090 individual college quarterbacks between the years of 2000 and 2019, that are distributed throughout the US as shown in the following map.

Figure 2. College location for 2010-2016 data

We also have the following characteristics for each of the players, as described in the table below:

Table 2: Variables description and distribution for College QBs by category (Source: sports-reference.com and ESPN)

Preliminary Results

High School Level

As a first approach, we will analyze the top quarterback recruits between 2000 and 2017, and follow their progression into college football. By focusing on the top recruit prospects (as defined by ESPN and 247Sports), we are able to analyze the top contenders for QB positions each year. In particular, we will focus on two specific questions:

1. Is there a disparity in recruitment by race?
2. If recruited, is there a shift in position that is different by race?

The first question addresses the issue of potential discrimination on the extensive margin, and the second one tackles the issue of potential discrimination in terms of the specific QB position.

In our data, 74% of top recruits end up accepting an offer from a college to play football, and there is no significant difference by race. Regarding shifts or changes in position, however, the story is very different. While the majority of top high school QBs are recruited to attend college to play football, only 80% of those recruited end up in the quarterback position in college and two thirds of those position shifts are for QBs of color.

The following table shows the comparison between white and non-white top recruits in terms of some of their characteristics. As show in this table, even though both groups are comparable in terms of performance ratings and physical attributes, ranked QBs of color have a much higher ranking than white quarterbacks. There is also a significant difference in how they are classified: Pocket passers or dual threats. Additionally, there is a significant difference in the probability of being recruited as a QB in college, but not for being recruited for college football.

Table 3: Comparison of characteristics between white vs non-white Top QB Recruits (2006-2017) (Source: ESPN and 247Sports)

	White (N=1392)		Non-white (N=748)
	Mean	Std. Dev.	Mean	Std. Dev.	Diff. in Means	p
Year	2009.70	4.50	2010.47	4.60	0.77	<0.01
Ranking (ESPN)	45.98	28.02	39.00	24.62	-6.98	<0.01
Grade (ESPN)	76.05	4.57	76.14	4.44	0.09	0.73
Ranking (247)	24.39	13.92	22.79	14.23	-1.60	0.03
Rating (247)	0.86	0.06	0.86	0.06	0.00	0.50
Height (ft)	6.22	0.13	6.16	0.15	-0.06	<0.01
Weight (lbs)	198.01	14.72	194.31	15.91	-3.70	<0.01
Dual Threat List	0.27	0.44	0.71	0.46	0.44	<0.01
Pocket Passer/Pro-Style List	0.60	0.49	0.18	0.39	-0.42	<0.01
Not classified	0.13	0.34	0.11	0.32	-0.02	0.23
College Recruit	0.74	0.44	0.72	0.45	-0.02	0.27
Recruited for College as QB	0.64	0.48	0.51	0.50	-0.13	<0.01
Recruited for College not as QB	0.11	0.31	0.21	0.41	0.11	<0.01

We ran five different models to analyze the difference in probability of being recruited as a college QB by race. Results show that high school quarterbacks of color are, on average, 13 percentage points less likely to be recruited as a QB for college football, even after controlling for performance and physical characteristics using different measures¹. We also included their play style classification (dual threat or pocket passers) for QBs and the difference in probability still remains. These players will then move on to other positions during their college years, which increases the disparities in the next selection stage, the NFL draft.

Table 4: Linear Probability Models for being recruited as a QB in College as a Top HS QB, based on race and performance

	College QB	College QB	College QB	College QB	College QB
NonWhiteQB	−0.133***	−0.134***	−0.114**	−0.127***	−0.127***
	(0.022)	(0.025)	(0.044)	(0.027)	(0.027)
Dual Threat List		0.003	0.016	−0.028	−0.036
		(0.024)	(0.043)	(0.026)	(0.027)
Grade (ESPN)			0.023***
			(0.003)
Rating (247)				2.774***	2.857***
				(0.178)	(0.187)
Height					−0.006
					(0.099)
Weight					−0.001+
					(0.001)
N	2140	2140	686	1602	1600
Avg. Outcome (Control)	0.639	0.639	0.644	0.722	0.722
Year FE	Yes	Yes	Yes	Yes	Yes
Years	00-17	00-17	13-17	00-17	00-17
R2 Adj.	0.028	0.028	0.101	0.137	0.138
+ p < 0.1, * p < 0.05, ** p < 0.01, *** p < 0.001
Robust SE in parenthesis.

In order to analyze the heterogeneity across years, we ran the previous regression (column 5) for different years. The following figure shows the estimated coefficients for NonWhiteQB for all years in our data, starting in 2002. As a reference, 2011 was the year Cam Newton was drafted into the NFL. Each yearly estimate is noisy, given that there are between 100 to 150 total observations per year, but there is a slight decrease in the magnitude of the estimate throughout the years.

Figure 3. Difference in probability of QB college recruitment for non-white HS QBs compared to white HS QBs

Given that white and non-white QBs could also have different styles of play, where one could be better rewarded because of their style, we re-weight non-white quarterback observations to match the characteristics of white QBs in terms of overall scouts’ grade, style of play (e.g. pocket passers or dual threats), weight, height, and year of graduation.

We follow Chattopadhyay, Hase, and Zubizarreta (2020) for building stable weights that approximately balance the empirical distribution of the aforementioned covariates with a tolerance of 0.01 SD, using the distribution of performance for white high school QBs as the target. The following table shows the balance between both groups after weighting.

Table 5: Average charactersitics between white and non-white HS QBs after Stable Balance Weighting

	White HS QB	Non-white HS QB	Target	Diff in Means (SD)
Year	2009.68	2009.73	2009.68	0.010
Espn data	0.67	0.66	0.67	0.010
247Sports data	0.74	0.74	0.74	0.010
Grade (ESPN)	59.99	60.28	59.99	0.010
Rating (247)	0.86	0.86	0.86	0.010
Weight (lbs)	197.98	198.04	197.98	0.004
Height (ft)	6.22	6.22	6.22	0.010
Dual threat list	0.27	0.27	0.27	0.010
Pocket passer list	0.60	0.60	0.60	0.010

Comparing the probability of being recruited to play in college as a QB for both groups after weighting, we find that there is a 0.11 (SE = 0.014; p-value = 0) decrease in the average probability of being recruited for high school QBs of color compared to white QBs, which is comparable in magnitude to our previous result for the same time period.

In a similar exercise, we now directly match white HS QBs to QBs of color for each year using the characteristics available. The idea is to create matched pairs of the same style of players (i.e. Pro Style or Dual Threats), with similar characteristics of performance (rating and ranking), as well as physical attributes (height and weight), where the pairs only differ by race. The matched groups in this case have the following characteristics²:

Table 6: Balance table for high school statistics and physical measures after matching (2000-2017) (Source: 247Sports and ESPN)

	Non-white QB (N=478)		White QB (N=478)
	Mean	Std. Dev.	Mean	Std. Dev.	Diff. in Means	p
Height (ft)	6.18	0.14	6.18	0.14	0.00	0.92
Weight (lbs)	195.41	16.30	195.47	14.13	0.07	0.94
Ranking (ESPN)	38.23	24.29	38.27	23.36	0.04	0.98
Grade (ESPN)	76.12	4.73	76.17	4.70	0.05	0.91
Ranking (247Sports)	23.95	14.22	24.06	13.61	0.11	0.91
Rating (247Sports)	0.85	0.06	0.85	0.06	0.00	0.81
Dual Threat (247Sports)	0.72	0.45	0.72	0.45	0.00	1.00
Year	2009.87	4.80	2009.87	4.80	0.00	1.00
N offers (247Sports)	3.26	1.74	3.25	1.74	-0.01	0.94

In this case, the difference in probability between matched groups is -0.138 (95% CI = [-0.195,-0.081]), which means that QBs of color, after matching by play style classification, performance, physical characteristics, overall interests from schools, and year, are 13.8 percentage points less likely to be recruited as a college QB than a white high school quarterback. There is also a 0.069 (95% CI = [0.023,0.115]) increase in probability for black QBs of being recruited in a different position for college.

The advantage of matching units is that we can directly compare both groups without using extrapolation or relying on parametric functional form assumptions. Additionally, this approach lends itself nicely to simple and straightforward sensitivity analysis to hidden bias (Rosenbaum 2002). The idea behind this type of sensitivity analysis relies on the fact that, within a matched set in a randomized experiment, each unit has the same probability of assignment to treatment (50% in a two-way matching design). In an observational study, however, there can be unobserved confounding that skews these probabilities. In this case, for a matched pair of QBs of different race, we will assume that one unit is \(\Gamma>1\) times more likely to be recruited as a college QB based on unobserved characteristics, and those characteristics can explain away any significant differences we find.

For this specific design, we find a sensitivity to hidden bias of \(\Gamma =\) 1.65, which means that an unobserved confounder should change the probability of assignment from 50%-50% to 37.7% - 62.3% in order to qualitatively explain away our findings, which implies that our estimate is moderately robust to hidden bias, according to literature parameters. To put this measure in context of other characteristics, the prior result would be equivalent to the difference in QB recruitment for two matched groups that are balanced in expectation on all characteristics (observed and unobserved), with the exception of an imbalance of 0.56 SD in rating score (247 Sports). This means that an unobserved confounder that has not been accounted for would have to have the same effect as almost 0.6 SD difference in HS QB performance based on scouts ratings.

Dual Threats vs Pocket Passers

Quarterbacks are usually categorized as either dual threats or pocket passers according to their style of play. QBs categorized as dual threats are considered players who have the skills and physical attributes to run with the ball, if needed. Pocket passers (or pro-style), on the other hand, are players that are able to generate plays and passes within the “pocket”, the area created by the blockers in an offensive play. The latter style of play has been the predominant one in professional football, at least until 2010.

Historically, “dual threat” has been a racially coded term usually assigned to Black QBs, and, given the playing style in the professional league, could be a reason for why QBs of color have lower recruitment probabilities as quarterbacks or get shifted to a different position in their transition to college. To analyze this hypothesis, we first analyze the probability of being categorized as a dual threat vs a pocket passer for top high school QBs between 2000 and 2017 using categorization from Sports 247 and ESPN³. As seen in the following table, QBs of color are much more likely to be labeled a dual threat vs a pocket passer, even after controlling for some physical characteristics and their grade performance.

Table 7: Linear Probability Models for being categorized as a Dual Threat (vs a Pocket Passer) as a Top HS QB, based on race and performance

	Dual Threat	Dual Threat	Dual Threat
NonWhiteQB	0.426***	0.494***	0.419***
	(0.021)	(0.035)	(0.023)
Height	−0.890***	−0.908***	−0.833***
	(0.076)	(0.122)	(0.086)
Weight	−0.002**	−0.001	−0.002*
	(0.001)	(0.001)	(0.001)
Grade		0.005+
		(0.003)
Rating (247)			−1.031***
			(0.203)
N	1873	686	1600
Avg. Outcome (Control)	0.271	0.26	0.325
Year FE	Yes	Yes	Yes
Years	00-17	13-17	00-17
R2 Adj.	0.301	0.386	0.301
+ p < 0.1, * p < 0.05, ** p < 0.01, *** p < 0.001
Robust SE in parenthesis.

Looking now at the association between their high school categorization and their performance in college, we can see that players that were categorized as dual threats do make, on average, fewer passing attempts in a year compared to those labeled as pocket passers. However, their effectiveness in passing plays is similar, with no statistically significant difference in average passing yards and effective passing rates. However, as expected, they compensate with a higher number of rushing attempts and average rushing yards, showing no statistically significance in the total number of touch downs contributed, either by passing or rushing plays, or their quarterback rating⁴.

Table 8: Linear Models of Passing Statistics for College QBs, based on high school label, race, and high school performance

	Passing Attempts	Avg Passing Yards	Eff. Passing Rate
Dual Threat	−36.172***	−0.066	−2.215
	(4.928)	(0.125)	(2.046)
Games Played	32.443***	0.156***	2.914***
	(0.550)	(0.017)	(0.285)
Height	41.909*	−0.399	−11.589
	(20.765)	(0.513)	(8.770)
Weight	−0.528**	−0.005	−0.079
	(0.181)	(0.004)	(0.071)
Rating (247)	46.341	3.200***	42.436**
	(40.743)	(0.968)	(14.645)
N	2867	2810	2810
Avg. Outcome (Control)	165.392	6.68	120.981
Year FE	Yes	Yes	Yes
Years	00-19	00-19	00-19
R2 Adj.	0.649	0.055	0.065
+ p < 0.1, * p < 0.05, ** p < 0.01, *** p < 0.001
Clustered robust SE in parenthesis (clustered at the player level).

Table 9: Linear Models of Rushing Statistics and Overall Performance for College QBs, based on high school label, race, and high school performance

	Rushing Attempts	Avg Rushing Yards	Total TDs	QBR
Dual Threat	18.804***	2.458***	−0.187	1.366
	(1.940)	(0.171)	(0.406)	(1.141)
Games Played	7.968***	0.056**	2.204***	3.681***
	(0.220)	(0.019)	(0.048)	(0.296)
Height	−12.771	−1.381+	1.575	2.332
	(8.229)	(0.725)	(1.762)	(5.045)
Weight	0.145*	0.001	−0.015	−0.055
	(0.067)	(0.006)	(0.014)	(0.042)
Rating (247)	−17.531	−0.430	9.868**	59.784***
	(16.754)	(1.409)	(3.258)	(9.140)
N	2905	2905	2905	1134
Avg. Outcome (Control)	35.7	0.929	9.358	56.793
Year FE	Yes	Yes	Yes	Yes
Years	00-19	00-19	00-19	04-19
R2 Adj.	0.499	0.139	0.580	0.253
+ p < 0.1, * p < 0.05, ** p < 0.01, *** p < 0.001
Clustered robust SE in parenthesis (clustered at the player level).

Finally, we analyze the probability of being shifted to a different position based on the high school label provided for players that are recruited to college football. On average, high school QBs categorized as dual threats are 10 percentage points more likely to switch positions in college compared to pocket passers, who, on average, only switch positions 10% of the time.

Table 10: Linear Probability Models for switching positions in college as a Top 100 HS QB, based on label, race and performance

	Switch Position	Switch Position	Switch Position	Switch Position	Switch Position
Dual Threat	0.141***	0.104***	0.110***		0.075**
	(0.020)	(0.020)	(0.020)		(0.026)
Rating (247)		−1.370***	−1.443***	−1.591***	−1.525***
		(0.168)	(0.178)	(0.178)	(0.178)
Height			−0.013	−0.046	0.004
			(0.099)	(0.096)	(0.099)
Weight			0.001	0.001	0.001
			(0.001)	(0.001)	(0.001)
NonWhiteQB				0.147***	0.162***
				(0.023)	(0.044)
Dual Threat × NonWhiteQB					−0.061
					(0.055)
N	1452	1272	1271	1271	1271
Avg. Outcome (Control)	0.11	0.095	0.095	0.095	0.095
Year FE	Yes	Yes	Yes	Yes	Yes
Years	00-17	00-17	00-17	00-17	00-17
R2 Adj.	0.038	0.077	0.078	0.094	0.098
+ p < 0.1, * p < 0.05, ** p < 0.01, *** p < 0.001
Robust SE in parenthesis.

College Level

After analyzing the leaks in the pipeline between high school and college, we will now move on to analyzing players during their college years. In this case, we will compare the performance in college football between white quarterbacks and quarterbacks of color. One key difference between both groups is that non-white QBs have significantly more rushing yards than white QBs, while less passing attempts. Both groups perform similarly, though, in terms of average passing yards, rate, and touchdowns. Non-white QBs also show a slight higher QB rating, as measured by ESPN for top QBs each year. In terms of drafting prospects, while both groups are equally likely to be drafted, non-white QBs have a lower probability of being drafted as a QB to the NFL.

Table 11: Comparison of characteristics between white vs non-white College QB (2006-2019) (Source: sports-reference.com and ESPN)

	White (N=5054)		Non-white (N=2248)
	Mean	Std. Dev.	Mean	Std. Dev.	Diff. in Means	p
Games Played	7.57	4.09	8.31	3.90	0.73	<0.01
Attempts (Pass)	142.27	158.20	120.53	140.05	-21.74	<0.01
Avg Yards Yards (Pass)	6.64	3.78	6.90	4.62	0.26	0.03
Rate (Pass)	119.30	67.05	121.70	79.93	2.40	0.26
Interceptions (Pass)	4.26	4.70	3.63	4.20	-0.63	<0.01
Attempts (Rush)	34.89	41.30	51.70	55.22	16.80	<0.01
Avg Yards (Rush)	1.60	4.60	3.82	4.08	2.23	<0.01
TDs	8.52	10.73	8.83	10.80	0.31	0.29
QBR	54.98	16.54	56.92	17.23	1.94	0.02
Drafted	0.13	0.33	0.12	0.33	0.00	0.55
Drafted as QB	0.09	0.29	0.06	0.23	-0.03	<0.01
Pick (Draft)	126.60	79.95	117.42	78.56	-9.18	0.11
Round (Draft)	4.18	2.12	3.89	2.08	-0.30	0.05

Figure 4. SD Difference in QB charactersitics by race (non-white vs white)

As it was seen in the previous table, in terms of their professional prospects measured as being drafted by an NFL team as a quarterback, results differ between both groups. The following models shows the difference in probability of being drafted (NonWhiteQB) for non-white QBs compared to white QBs. Even after controlling for year and some performance metrics, such as QB rating and passing touch downs, the differences persist:

Table 12: Linear Probability Models for being drafted as a QB for College QBs, based on race and performance

	Drafted	Drafted	Drafted	Drafted
NonWhiteQB	−0.033**	−0.043***	−0.044***	−0.056**
	(0.010)	(0.011)	(0.012)	(0.021)
Plays		0.012***	0.013***	0.011+
		(0.001)	(0.001)	(0.006)
Pass			0.001*	−0.012
			(0.001)	(0.011)
QBR				0.005***
				(0.001)
N	7302	6774	6240	1901
Avg. Outcome (Control)	0.09	0.09	0.09	0.171
Year FE	Yes	Yes	Yes	Yes
R2 Adj.	0.006	0.119	0.039	0.052
+ p < 0.1, * p < 0.05, ** p < 0.01, *** p < 0.001
Clustered robust SE in parenthesis (clustered at the player level).
QB rating (QBR) only available for top 120 quarterbacks each year, since 2004.

On average, a white college QB has a 17% probability of being drafted in our dataset⁵, which means that a negative estimate between 4.3 and 5.6 percentage points on being drafted represents a decrease in probability between 25% and 33% associated to being a QB of color.

In order to match the play style between white QBs and QBs of color, we perform the same re-weighting exercised as before, matching the style of play of white QBs in terms of passing yards/attempts, rushing yards/attempts, interceptions, touch downs, games played, and other aggregated performance characteristics such as QB Rating, when available.

The following table shows the balance between both groups after weighting.

Table 13: Average charactersitics between white and non-white QBs after Stable Balance Weighting

	White QB	Non-white QB	Target	Diff in Means (SD)
Games	11.52	11.53	11.52	0.01
Attempts (Pass)	340.81	339.76	340.81	0.01
Avg Yards (Pass)	7.27	7.28	7.27	0.01
Attempts (Rush)	75.46	75.98	75.46	0.01
Rate (Pass)	132.09	131.89	132.09	0.01
Avg Yards (Rush)	0.94	0.97	0.94	0.01
Interceptions (Pass)	9.39	9.35	9.39	0.01
TDs (Pass)	17.74	17.83	17.74	0.01
TDs (Rush)	3.09	3.04	3.09	0.01
QBR	54.99	54.82	54.99	0.01

After weighting for style of play, we find that there is a 0.047 (SE = 0.011; p-value = 0) decrease in the average probability of being drafted for QBs of color compared to white QBs. This difference aligns with our previous results, even when matching the style of play of black QBs to the style of white QBs in college. These findings indicate that even though there could be other potential confounders in play, playing style as captured by these main metrics are not the ones driving this significant difference in drafting probability.

Heterogenous effects

We are also interested in analyzing whether there are differences in recruitment patterns by state, particularly focused on how conservative or liberal these places are. As a proxy for how conservative a state is, we use the results for the presidential elections between 2000 and 2020 and the proportion of votes the republican candidate obtained in the ballots⁶. Additionally, we collect information from 274sports.com about colleges that were interested in a specific player, whether they made an offer, and where the player ended up accepting an offer in the end.

Figure 3. Difference in probability of QB college recruitment for non-white HS QBs compared to white HS QBs

Footnotes

1. We use the performance metrics provided both by ESPN (since 2006 onwards) as well as 247Sports as an overall measure of HS performance.

2. We match exactly by style of play, and mean balance performance and physical attribute variables directly with a restriction of 0.025 SD. Finally, we fine balance for year of graduation.

3. ESPN only classified players as dual threats and pocket passers since 2013 onward.

4. This is a rating score designed by ESPN to rank QBs, and it has been assigned to the top 120 QBs in college since 2004.

5. This result is for the top 120 ranked QBs in college

6. Because these elections are only held every four years, we assign the result of the closest election year as a measure of how conservative a state is.